jameshartig commented on code in PR #1793:
URL:
https://github.com/apache/cassandra-gocql-driver/pull/1793#discussion_r1937461606
##########
query_executor.go:
##########
@@ -83,7 +84,27 @@ func (q *queryExecutor) speculate(ctx context.Context, qry
ExecutableQuery, sp S
}
func (q *queryExecutor) executeQuery(qry ExecutableQuery) (*Iter, error) {
- hostIter := q.policy.Pick(qry)
+ var hostIter NextHost
+
+ // check if the host id is specified for the query,
+ // if it is, the query should be executed at the corresponding host.
+ if hostID := qry.GetHostID(); hostID != "" {
+ pool, ok := q.pool.getPoolByHostID(hostID)
+ if !ok || !pool.host.IsUp() {
+ return &Iter{err: ErrNoConnections}, nil
+ }
+ hostIter = func() SelectedHost {
+ // forcing hostIter to always return the same host
+ // it makes any retries and speculative executions run
on the specified host
Review Comment:
> On line
[158](https://github.com/apache/cassandra-gocql-driver/blob/4a3a53b65f0726bc30fe14bac9be9baa8497c8ea/query_executor.go#L158)
will continue loop iteration if the specified host is down, so it never
reaches
[179](https://github.com/apache/cassandra-gocql-driver/blob/4a3a53b65f0726bc30fe14bac9be9baa8497c8ea/query_executor.go#L179).
Can you file that as a bug? That could affect the normal flow as well
separate from this. We should be tracking the last host and maybe erroring if
we get the same host twice. I don't think we need to alter this just because of
that specific case.
> Right, if we do this way retries will be disabled
I still think we should be aiming to make retries work, even if its in a
follow-up issue. There are transient things that happen in the real world that
might be annoying to handle if we just always disable retries.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
